Research report
Concordance between the PHQ-9 and the HSCL-20 in depressed primary care patients

https://doi.org/10.1016/j.jad.2006.09.002Get rights and content

Abstract

Background

Two instruments commonly used in primary care research to measure depressive severity are the Patient Health Questionnaire-9 (PHQ-9) and the Hopkins Symptom Checklist-20 (HSCL-20). However, there is little information regarding the relationship between clinical information derived from these scales. The present study investigates the psychometric properties of the PHQ-9 and HSCL-20, determines the degree of instrument concordance, and describes the factor structure of the HSCL-20.

Methods

A secondary data analysis from a randomized controlled trial was performed. A total of 405 primary care patients with major depressive disorder and/or dysthymia were administered the PHQ-9 and the HSCL-20 when recruited for the study.

Results

Good internal consistency reliability estimates were obtained for both scales (PHQ-9 α = 0.803; HSCL-20 α = 0.837). All PHQ-9 inter-item and corrected item–total correlations showed that no item detracted from overall scale functioning. HSCL-20 items assessing overeating, poor appetite, and sexual interest were poorly correlated with other items and with the total scale score. A positive, moderate strength relationship was found between the instruments (r = 0.54, p < 0.0001). Exploratory factor analysis of the HSCL-20 yielded a six-factor structure, which accounted for almost 63% of the variance in total score. The largest contribution to common variance in the scale was provided by an “anxiety and self-reproach” factor.

Conclusions

PHQ-9 and HSCL-20 total scores were moderately correlated. Although the HSCL-20 is utilized as a measure of depression severity, it may lack sufficient specificity to be an accurate reflection of depression status per se.

Introduction

Primary care clinicians can play an important role in the treatment of depression. However, detection of major depressive disorder in primary care patients is often poor (Davidson and Meltzer-Brody, 1999, Garrard et al., 1998, Rost et al., 1998, Spitzer et al., 1994), and once detected, treatment and monitoring often fall short of recommended guidelines (Simon, 1998). The use of accurate and efficient screening and symptom monitoring instruments may be one way to improve treatment of depression in primary care patients (Löwe et al., 2004).

Two depression instruments commonly used in primary care research are the Patient Health Questionnaire-9 (PHQ-9) (Spitzer et al., 1999, Kroenke et al., 2001a) and the Hopkins Symptom Checklist Depression Scale-20 (HSCL-20) (Derogatis et al., 1973). The PHQ-9 includes nine questions that correspond to the DSM-IV (American Psychiatric Association, 1994) symptoms of depression. A straightforward algorithm can be applied to determine the possible presence of a depressive disorder. When used to screen for depression, the PHQ-9 has shown strong psychometric properties including high internal consistency, face validity, test–retest reliability, and concurrent validity with measures of functional impairment (Corson et al., 2004, Kroenke et al., 2001a). The PHQ-9 algorithm for major depression has shown good sensitivity (73%–91%) and high specificity (89%–94%) when compared to a mental health professional interview (Spitzer et al., 1999, Williams et al., 2005).

In addition to its use as a screening tool, the PHQ-9 has demonstrated validity when used to measure depressive severity (Kroenke et al., 2001a). Importantly, the PHQ-9 has also been shown to be sensitive to change in patients being treated for depression. Currently, the PHQ-9 is the only self-report depression instrument that has been validated as a screening, severity and outcome measure (Löwe et al., 2004). Given its desirable psychometric attributes, brevity, and ease of administration and scoring, the PHQ-9 is a versatile tool for use in the busy primary care setting.

The Hopkins Symptom Checklist Depression Scale-20 item version (HSCL-20) is a self-report scale comprised of the 13 items of the Hopkins Symptom Checklist Depression Scale (HSCL-D) plus 7 additional items from the Hopkins Symptom Checklist-90-Revised (HSCL-90-R). The additional 7 items were added so as to better represent all diagnostic symptoms of major depression and improve the instrument's sensitivity to clinical change (Williams et al., 2004).

Although the HSCL-20 has been widely used as a measure of depressive severity in large clinical trials (Boudreau et al., 2002, Felker et al., 2001, Fraser et al., 2004, Hedrick et al., 2003, Katon et al., 1996, Kroenke et al., 2001b, Unützer et al., 2002, Williams et al., 2000) several psychometric characteristics of the HSCL-20, including criterion validity, test–retest reliability, and factor structure remain unexplored.

Given the availability of these two instruments, how comparable is the clinical information generated by each of them? Few studies have compared these commonly utilized depression inventories. Löwe et al. (2004) found that when compared to the HSCL-20, PHQ-9 responsiveness was significantly greater at 3 months and equivalent at 6 months. The purpose of the present study, therefore, is to advance knowledge of these instruments by: 1) investigating the psychometric properties (item and scale reliabilities) of the HSCL-20 and PHQ-9 in a sample of depressed primary care patients; 2) determining the degree of concordance between the HSCL-20 and PHQ-9; and 3) describing the factor structure of the HSCL-20 through exploratory factor analysis.

Section snippets

Methods

Characteristics of the PHQ-9 and HSCL-20 were explored as part of a randomized controlled trial assessing the impact of the Three Component Model (TCM) of depression care (Dietrich et al., 2004a). The research protocol was approved by the Committees for Human Subjects Protection at the participating healthcare and research organizations. The study methods, summarized here, are presented in more detail elsewhere (Dietrich et al., 2004b).

Participants, aged 18 years or older, were identified as

Results

Primary care clinicians referred 987 possibly depressed patients for study participation. Of these, 861 completed the PHQ-9 screening and 405 were eligible for study participation and completed the baseline interview. The mean age of participants was 41.9 years (SD 14.6) and ranged from 18 to 94 years. The majority of participants were female (80.3%), white (83.1%), married (55.3%) and employed (61.5%). The average education level reported by study participants was 13.2 years. Approximately 20%

Discussion

In the present study of the psychometric properties of two depression instruments, both the HSCL-20 and PHQ-9 demonstrated good internal consistency when administered to primary care patients.

All PHQ-9 inter-item correlations and corrected item–total correlations demonstrated that no single item detracted in any significant way from overall scale functioning. This was not true, however, for the HSCL-20 wherein overeating, poor appetite, and sexual interest were inadequately correlated with

Acknowledgments

This research was supported by the John D. and Catherine T. MacArthur Foundation.

This analysis was submitted in partial fulfillment of Pamela W. Lee's doctoral dissertation requirements. Special thanks are extended to Dr. Charles Swencionis and Dr. Jonathan Feldman of the Ferkauf Graduate School of Psychology of Yeshiva University, Bronx, NY and Dr. Ellen Brown and Dr. Christopher Murphy of the Weill Medical College of Cornell University, White Plains, NY. Their guidance is gratefully

References (31)

  • A.J. Dietrich et al.

    Going to scale: re-engineering systems for primary care treatment of depression

    Ann. Fam. Med.

    (2004)
  • J. Garrard et al.

    Clinical detection of depression among community-based elderly people with self-reported symptoms of depression

    J. Gerontol., Ser. A, Biol. Sci. Med. Sci.

    (1998)
  • S.C. Hedrick et al.

    Effectiveness of collaborative care depression treatment in Veterans' Affairs primary care

    J. Gen. Intern. Med.

    (2003)
  • R.M. Hirschfeld

    The comorbidity of major depression and anxiety disorders: recognition and management in primary care

    Prim. Care Companion J. Clin. Psychiat.

    (2001)
  • F.Y. Huang et al.

    Using the PHQ-9 to measure depression among racially and ethnically diverse primary care patients

    J. Gen. Intern. Med.

    (2006)
  • Cited by (69)

    • Is improvement in comorbid major depression associated with longer survival in people with cancer? A long-term follow-up of participants in the SMaRT Oncology-2 and 3 trials

      2019, Journal of Psychosomatic Research
      Citation Excerpt :

      A one unit decrease in the overall score indicates a mean decrease of one level of severity across the 20 symptoms. The scale has been found to have good agreement with other commonly used measures of depression severity, such as the Patient Health Questionnaire-9 and the Hamilton Depression Rating Scale [13,20]. For this analysis, we defined change in depression severity as the change in a participant's SCL-20 score between baseline (trial randomisation) and 12-weeks post-randomisation.

    View all citing articles on Scopus
    View full text